Animal Models of Diabetic Neuropathy: Progress Since 1960s

Abstract

Diabetic or peripheral diabetic neuropathy (PDN) is one of the major complications among some other diabetic complications such as diabetic nephropathy, diabetic retinopathy, and diabetic cardiomyopathy. The use of animal models in the research of diabetes and diabetic complications is very common when rats and mice are most commonly used for many reasons. A numbers of animal models of diabetic and PDN have been developed in the last several decades such as streptozotocin-induced diabetic rat models, conventional or genetically modified or high-fat diet-fed C57BL/Ks (db/db) mice models, streptozotocin-induced C57BL6/J and ddY mice models, Chinese hamster neuropathic model, rhesus monkey PDN model, spontaneously diabetic WBN/Kob rat model, L-fucose-induced neropathic rat model, partial sciatic nerve ligated rat model, nonobese diabetic (NOD) mice model, spontaneously induced Ins2 Akita mice model, leptin-deficient (ob/ob) mice model, Otsuka Long-Evans Tokushima Fatty (OLETF) rat model, surgically-induced neuropathic model, and genetically modified Spontaneously Diabetic Torii (SDT) rat model, none of which are without limitations. An animal model of diabetic or PDN should mimic the all major pathogeneses of human diabetic neuropathy. Hence, this review comparatively evaluates the animal models of diabetic and PDN which are developed since 1960s with their advantages and disadvantages to help diabetic research groups in order to more accurately choose an appropriate model to meet their specific research objectives.

Full text

Hindawi Publishing Corporation
Journal of Diabetes Research
Volume , Article ID ,  pages
http://dx.doi.org/.//
Review Article
Animal Models of Diabetic Neuropathy: Progress Since 1960s
Md. Shahidul Islam
Department of Biochemistry, School of Life Sciences, University of KwaZulu-Natal (Westville Campus), Durban 4000, South Africa
Correspondence should be addressed to Md. Shahidul Islam; islamd@ukzn.ac.za
Received  May ; Accepted  July 
Academic Editor: Daisuke Koya
Copyright ©  Md. Shahidul Islam. is is an open access article distributed under the Creative Commons Attribution License,
which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Diabetic or peripheral diabetic neuropathy (PDN) is one of the major complications among some other diabetic complications such
as diabetic nephropathy, diabetic retinopathy, and diabetic cardiomyopathy. e use of animal models in the research of diabetes
and diabetic complications is very common when rats and mice are most commonly used for many reasons. A numbers of animal
models of diabetic and PDN have been developed in the last several decades such as streptozotocin-induced diabetic rat models,
conventional or genetically modied or high-fat diet-fed CBL/Ks (db/db) mice models, streptozotocin-induced CBL/J and
ddY mice models, Chinese hamster neuropathic model, rhesus monkey PDN model, spontaneously diabetic WBN/Kob rat
model, L-fucose-induced neropathic rat model, partial sciatic nerve ligated rat model, nonobese diabetic (NOD) mice model,
spontaneously induced Ins Akita mice model, leptin-decient (ob/ob) mice model, Otsuka Long-Evans Tokushima Fatty (OLETF)
rat model, surgically-induced neuropathic model, and genetically modied Spontaneously Diabetic Torii (SDT) rat model, none of
which are without limitations. An animal model of diabetic or PDN should mimic the all major pathogeneses of human diabetic
neuropathy. Hence, this review comparatively evaluates the animal models of diabetic and PDN which are developed since s
with their advantages and disadvantages to help diabetic research groups in order to more accurately choose an appropriate model
to meet their specic research objectives.
1. Introduction
e term “diabetes” was rst coined by Araetus of Cap-
podocia (-AD). Later, the word “mellitus” (honey sweet)
wasaddedbyomasWillis(Britain)inaerrediscover-
ing the sweetness of urine and blood of patients (rst noticed
by the ancient Indians) []. In , Dobson (Britain) for the
rst time conrmed the presence of excess sugar in urine and
blood as a cause of their sweetness. Depending on the patho-
genesis, diabetes is classied as type  and type . e rst
widely accepted classication of diabetes mellitus was pub-
lished by World Health Organization (WHO) in  []
and, in modied form, in  []. In , the WHO Expert
Committee proposed two major classes of diabetes melli-
tus, namely: Insulin Dependent Diabetes Mellitus (IDDM)
or Type  and Noninsulin Dependent Diabetes Mellitus
(NIDDM) or Type  diabetes (TD). In , the WHO expert
committee omitted the terms Type  and Type , but the
terms IDDM and NIDDM were retained, and a class of
Malnutrition-Related Diabetes Mellitus (MRDM) was intro-
duced []. In both reports ( and ), other classes of
diabetes were also included, for example, Impaired Glucose
Tolerance (IGT) and Gestational Diabetes Mellitus (GDM)
[, ]. ese were reected in the subsequent International
Nomenclature of Disease (IND) in  and in the tenth revi-
sion of the International Classication of Diseases (ICD-)
in . e  classication was widely accepted and used
internationally even today.
Since last few decades, diagnosis of diabetes is not only
limited in blood and urine sugar levels but also in many other
parameters and factors such as serum insulin levels, blood
glycated haemoglobin and proteins, glucose tolerance ability,
insulin sensitivity or insulin resistance, pancreatic beta-cell
function, and so forth. Apart from above-mentioned parame-
ters related abnormalities, diabetes patients are oen suered
from other diabetes related complications such as—diabetic
neuropathy, diabetic cardiomyopathy, diabetic nephropathy
(DN), and diabetic retinopathy. ese are usually caused by
the poor glycemic control or improper management of dia-
betes mellitus. About % of people with diabetes are aected
with one or more of the above complications. Amongst
others,diabeticneuropathyisonetheleadingandpainful

 Journal of Diabetes Research
complications usually suered by many diabetic patients;
however,thepathogenesisofthiscomplicationisstillnotfully
understoodduetotheabsenceofanauthenticanimalmodel
which fully mimics the complications of human diabetic neu-
ropathy.
Animal models in diabetes research are very common
when most of the existing models are developed as a conven-
tional model either for Type  or for TD. But very oen a
conventional model of diabetes cannot demonstrate the spe-
cic pathogenesis of diabetes related complications. ere-
fore, the necessity of the individual and specic model for
diabetic complications has been raised in the recent years to
achieve the authentic outcomes of specic research aims. A
number of animal models of diabetic neuropathy have been
developed in last few decades approaching from diverse point
of views. However, most of them did not receive much pop-
ularity because of their considerable number of limitations
and disadvantages. In a comprehensive review, Harati []
reported that the major handicap in studying diabetic neu-
ropathies is the lack of a suitable animal model that addresses
acute and chronic events leading to diabetic neuropathy.
Hence, in this review, the pathogenesis, advantages, disad-
vantages, and limitations of several genetic and nongenetic
animal models of diabetic neuropathy have been discussed
to substantiate their ecacy for human study and in order to
guide diabetes research groups to more accurately select the
most appropriate models to address their specic research
questions.
2. Animal Models in Diabetic Neuropathy
Peripheral diabetic neuropathy (PDN) is a shattering com-
plication of diabetes and leading cause of foot exclusion [].
Clinical indications of PDN include increased vibration and
thermal perception thresholds that progress to sensory loss,
occurring in conjunction with degeneration of all ber types
in the peripheral nerve []. A proportion of patients with
PDN also describe abnormal sensations such as paresthesia,
allodynia, hyperalgesia, and spontaneous pain that some-
times coexist with loss of normal sensory function [].
According to a recent review, a number of studies have inves-
tigated and described DN in mice, but it is dicult to compare
these studies with each other or with human DN due to
experimental dierences including the animal strain, type of
diabetes, method of induction, duration of diabetes, animal
age, and gender []. Although two review articles [, ]
on animal models of diabetic and some other neuropathies
are published recently, none of them suggested the most
suitable model in order to study the further pathogenesis of
diabetic neuropathy and also for the pharmacological screen-
ing and development of antidiabetic or anti-neuropathic
drug in their reviews. Shaikh and Somani [] simply
reviewed the behavioral, structural, functional, and molecu-
lar markers of Type  and Type  diabetic neurophaty while
H
¨
oke [] briey reviewed the physiological changes in
diabetic and some other peripheral neuropathies such as
chemotherapy-induced peripheral neuropathy and human-
immunodeciency virus-associated sensory neuropathies.
is review precisely discussed the progress with the animal
models of diabetic neuropathy which have been developed
in last few decades since early s with their advantages,
disadvantages, and limitations in order to assist scientists to
more appropriately choose a model based on their specic
research aims. Additionally, the characterization of neuropa-
thy or advantages and limitations or disadvantages of most of
the models are summarized in Table .
2.1. Models Developed during 1960s and 1970s. e nerve
conduction and regenerative changes in experimental dia-
betes were rst noticed by Eliasson during - [, ];
however, the rst peripheral neuropathy in alloxan-diabetic
rats was reported by Preston in  []thenLovelacein
 []. Aer that a number of scientists reported diabetic
neurophaty mostly in alloxan-induced diabetic models. A
complete animal (rat) model of diabetic neuropathy (DN)
was rst reported by Jakobsen and Lundbeck in  []with
reducedsizesofnerveber,axon,andmyelinsheath,which
contribute in impaired motor function in streptozotocin
(STZ)-induced diabetic rats. Aer a couple of years, during
–, animal model of PDN was rst reported as well
as evaluated by Sima and Robertson in several studies con-
ducted in streptozotocin-induced diabetic rats and mutant
diabetic [CBL/Ks (db/db)] mice [–]. e PDN was
initially characterized by severely decreased motor nerve
conduction velocity (MNCV), absence of large myelinated
bers, and axonal atrophy in this mouse model. In the further
evaluationstudies,axonalchangesaswellasaxonaldystrophy
wereobservedinthemyelinatedandunmyelinatedbers
followed by loss, shrinkage, and breakdown of myelin sheath
in the later stage. However, the major limitation is that none
of these models have been evaluated by using anti-diabetic or
antineuropathic drugs.
2.2. Models Developed during 1980s. In early s, PDN
wasassessedindiabeticChinesehamsterbyKennedyand
colleagues []. Conduction velocities in both motor and
sensory components of the hind lamb nerves were reduced
–% in diabetic compared to control animals. However,
there was no reduction in nerve ber diameters or other signs
of abnormal morphology that could be correlated with these
physiological eects. However, PDN in diabetic hamster is
less severe than human DN in its clinical stage. Hence, further
study is warranted to use this animal as a model for human
PDN. Cornblath et al. []triedtodevelopaprimatemodel
ofPDNinrhesusmonkey.eyfoundsignicantlyreduced
motor nerve conduction velocities and prolonged F-wave
latencies in diabetic animals compared to nondiabetic control
animals, while motor-evoked amplitudes did not dier. Addi-
tionally, nerve conduction times were increased in motor
bers of diabetic animals two years aer the onset of diabetic
hyperglycemia. Although these abnormalities are similar to
those seen in humans, further study is needed to establish this
primate model for human PDN since these models have not
been evaluated by any antineuropathic drugs. Additionally,
aer comparing with diabetic and hypoglycaemic neuropa-
thy, Sima et al. [] reported that diabetic neuropathy is not
associated with nerve cell loss but showed marked axonal

Journal of Diabetes Research 
T : Characterization criteria (advantages) and limitations (disadvantages) of some selective animal models of diabetic neuropathy
developed since s.
Animals models References
Characterization of diabetic
neuropathy/advantages
Limitations/disadvantages
Streptozotocin-
induced rat model
(classic)
Jakobsen and Lundbeck [].
(i) Reduced sizes of nerve ber, axon, and
myelin sheath.
(ii) Impaired motor function.
Not validated by
antineuropathic drug.
Streptozotocin-
induced rat model
(recent)
Filho and Fazan [].
(i) Signicantly reduced right and le fascicular
areas and myelination of phrenic nerves.
(ii) Validated by insulin (s.c.).
(i) Some major
pathogenesis of diabetic
neuropathy has not been
characterized.
(ii) Although validated by
insulin (s.c.), no
antineuropathic drug has
been used.
CBL/Ks (db/db)
mice model
(classic)
Sima and Robertson [, ];
Robertson and Sima [].
(i) Severely decreased motor nerve conduction
velocity (MNCV).
(ii) Absence of large myelinated bers.
(iii) Axonal atrophy.
(iv) Axonal dystrophy in myelinated and
unmyelinated bers.
(v) Loss, shrinkage, and breakdown of myeline
sheath.
Not evaluated by any
anti-diabetic or
antineuropathic drug.
Genetically
modied
CBL/Ks (db/db)
mice model
(recent)
Hinder et al. [].
(i) Increased body weight, hyperglycemia, and
hyperlipidemia.
(ii) Lower tail ick response to heat stimulus,
sciatic motor nerve conduction velocity, and
intraepididymal nerve ber velocity.
(i) Mismatched results were
observed for body weight,
blood glucose, plasma
lipids, and blood glycated
hemoglobin.
(ii) Not validated by
anti-diabetic or
antineuropathic drugs.
Streptozotocin-
induced CBL/J
mice model
Varen iuk et a l. [].
(i) Peroxynitrite injury in peripheral nerve and
dorsal root ganglion neurons.
(ii) Motor and sensory nerve conduction
velocity decits, thermal and mechanical
hyperplasia, tactile allodynia, and loss of
intraepidermal nerve bers.
Not validated by using
antineuropathic drug.
Streptozotocin-
induced diabetic
sensory
neuropathic ddY
mice model
Murakami et al. [].
(i) Signicantly lower sensory nerve
conduction velocity, higher nociceptive
threshold, hypoalgesia, and unmyelinated ber
atrophy.
(ii) Successfully evaluated by insulin treatment.
(iii) Can be a better model to study the human
sensory polyneuropathy.
No signicant change was
found in the myelinated
nerve ber areas.
Chinese hamster
neuropathic model
Kennedy et al. [].
Reduced conduction velocity of both motor
and sensory components of hind lamb nerves
(–%).
(i) Peripheral diabetic
neuropathy (PDN) was less
severe than human diabetic
neuropathy.
(ii) Further study needed
for proper validation.
Rhesus monkey
model of PDN
Cornblath et al. [].
(i) Signicantly reduced motor conduction
velocity.
(ii) Prolonged F-wave latencies.
(iii) Pathogeneses’ resembles to humans.
(i) No dierence in
motor-evoked amplitudes.
(ii) Prolonged nerve
conduction induction time
( years).
(iii) Not validated by
antineuropathic drug.

 Journal of Diabetes Research
T : C ontinued.
Animals models References
Characterization of diabetic
neuropathy/advantages
Limitations/disadvantages
Spontaneously
diabetic WBN/Kob
rat model
Yag ih a s h i et a l. [].
(i) Slower motor nerve conduction and
temporal dispersion of compound muscle
action potential.
(ii) Structural de- and remyelination in the
sciatic and tibial nerves at  month.
(iii) Axonal degeneration, dystrophy, and
reduced myelinated ber at  month.
(iv) Resembles human pathogenesis of PDN.
Not validated by
antineuropathic drug.
L-fucose induced
neuropathic rat
model
Sima et al. [].
(i) Reduced Na
+
-K
+
-ATPase activity.
(ii) Reduced nerve conduction velocity.
(iii) Axonal dystrophy.
(iv) Paranodal swelling and demyelination
without increasing Walleran degeneration of
nerve ber loss.
Not validated by
antineuropathic drug.
Partial sciatic nerve
ligated rat model
Fox et al. [].
(i) Produced long-lasting mechanical, but
thermal hyperalgesia.
(ii) Evaluated by ant-diabetic neuropathic
drugs.
Major pathogenesis was not
characterized.
Nonobese diabetic
(NOD) mice model
Schmidt et al. [];
Homs et al. [].
(i) Short induction period.
(ii) Markedly swollen axons and dendrites
(neurotic dystrophy).
(iii) Consistent with the pathogenesis of other
rodent models of PDN and human PDN.
(iv) Suggested as a better model than ICR mice
particularly in terms of nerve regeneration.
Not validated by
antineuropathic drug.
Spontaneously
induced Ins Akita
mouse model
Choeiri et al. [];
Schmidt et al. [].
(i) Spontaneously induced diabetic model.
(ii) Progressive and sustained chronic
hyperglycemia.
(iii) Reduced sensory nerve conduction
velocity.
(iv) Markedly swollen axons and dendrites
(neurotic dystrophy).
(v) Consistent with the pathogenesis of other
rodent models of PDN and human PDN.
Not validated by
anti-diabetic or
antineuropathic drug.
Leptin-decient
(ob/ob) mice
model
Drel et al. [].
(i) Clearly manifested thermal hypoalgesia. (ii)
Relatively higher nonfasting blood glucose
level ( mmol/L).
(iii) Slow motor and sensory nerve conduction.
(iv) Signicant reduction of intraepidermal
nerve ber.
(v) Validated by antiperipheral diabetic
neuropathic drug.
May not be widely available
for routine
pharmacological screening
of anti-diabetic or
anti-neuropathic drugs.
Otsuka
Long-Evans
Tokushima Fatty
(OLETF) rats
model
Kamenov et al. [].
(i) Signicantly higher blood glucose and
HbAc levels.
(ii) Reduced motor nerve conduction velocity
and thermal nociception.
(i) Some major
pathogenesis of PDN has
not been characterized.
(ii) Not validated by
anti-diabetic neuropathic
drugs.
Rat insulin I
promoter/human
interferon-beta
(RIP/IFN𝛽)
transgenic ICR
mice model
Seraf
´
ın et al. [].
(i) Signicantly hyperglycemia, slower tibial
sensory nerve conduction velocity.
(ii) Reduced nerve ber density and increased
motor latencies.
(i) A sophisticated surgical
approach has been used to
develop the model.
(ii) Not validated by
anti-diabetic or
antineuropathic drugs.

Journal of Diabetes Research 
T : C ontinued.
Animals models References
Characterization of diabetic
neuropathy/advantages
Limitations/disadvantages
High-fat diet-fed
female CBL/J
mice model
Obrosova et al. [].
(i) Decit of motor and sensory nerve
conductions, tactile allodynia, and thermal
hypoalgesia. (ii) Can be used as model for
prediabetic or obesity related neuropathy.
(i) Intradermal nerve ber
loss, and axonal atrophy
was absent.
(ii) Cannot be used for
chronic diabetic
neuropathy.
(iii) Not validated by
antineuropathic drugs.
Surgically-induced
neuropathic model
Muthuraman et al. [].
(i) ermal and mechanical hyperalgesia in
paw and tail.
(ii) Reduced nerve ber density and nerve
conduction velocity.
(iii) Very short induction period.
(i) Not validated by using
antineuropathic drug.
(ii) Not suitable to study
the human diabetic
neuropathy.
Genetically
modied SDT fatty
rat model
Yam a g u c h i et a l. [].
(i) Sustained hyperglycemia and dyslipidemia
with delayed and reduced motor nerve
conduction velocity.
(ii) Lower number of sural nerve bers and
thickened epinural arterioles.
(iii) Successfully validated by anti-diabetic
drug such as pioglitazone.
Some pathogenesis was
induced only aer a long
period of time such as 
weeks.
atrophy involving predominantly sensory bers. So this par-
ticular factor needs to be considered before choosing any ani-
mal model for a diabetic neuropathic study.
2.3. Models Developed during 1990s
2.3.1. Spontaneously Diabetic WBN/Kob Rat Model. In early
s, the model of PDN further developed in a sponta-
neously diabetic WBN/Kob rats via examining electrophys-
iologic, biochemical, and structural changes of peripheral
nervesatandmonthsofages[]. is model was char-
acterized by slower motor nerve conduction and temporal
dispersion of compound muscle action potential. Structural
de- and remyelinations were observed in the sciatic and tibial
nerves in -month-old rats, while -month-old rats addi-
tionally showed axonal degeneration and dystrophy, reduced
myelinated ber occupancy, and decreased mean myelinated
ber size. Additionally, these neuropathic manifestations are
unique as compared with those found in other spontaneously
diabetic animal models. is model of WBN/Kob rats is
further supported by Ozaki et al. [], because this model of
PDN develops primary segmental demyelination and sec-
ondary axonal degeneration, which are similar to those in
human patients with diabetes mellitus and unlike those in
rodents with streptozotocin-induced diabetes []. Hence,
spontaneously diabetic WBN/Kob rats can be a better model
to study the human PDN.
2.3.2. L-Fucose-Induced Rat Model. In late s, it has been
reported that L-fucose, a competitive inhibitor of sodium-
dependent myoinositol transport, has been shown eective
to induce diabetic neuropathy in normal rats mediated by
Na
+
-K
+
-ATPase activity and conduction of nerve velocity
[]. To further validate, long-term feeding of L-fucose has
beenstudiedinthismodelandevaluatedbynerveNa
+
-K
+
-
ATPase activity, conduction velocity, and myelinated nerve
ber pathology. Aer -week supplementation of L-fucose
enriched ( or %) diets, Na
+
-K
+
-ATPase activity was
signicantly decreased, associated with a –% reduction
in nerve conduction velocity. Twenty percent L-fucose diet
resulted in signicant axonal atrophy, paranodal swelling,
and paranodal demyelination without increasing Walleran
degeneration or nerve ber loss. Aer this study, it has
been recommended that this L-fucose model can serve as an
experimental tool to study the diabetic neuropathy.
2.3.3. Partial Sciatic-Nerve Ligated Rat Model. In another
study, partial ligation of sciatic nerve method has been used
to induce PDN and compared with a usual STZ-induced
rat model of PDN []. STZ-induced diabetic animals were
chronically ill, with reduced growth rate, polyuria, diarrhoea,
and enlarged and distended bladders when these symptoms
were not found in sciatic nerve ligated model. is sciatic
nerve ligated model has also been evaluated with antineu-
ropathic drugs (Morphine and L-Baclofen), which produce
greaterreversalofmechanicalhyperalgesiafollowingpartial
nerve ligation. ey also added that STZ-induced diabetes in
rats produces long-lasting mechanical but not thermal hyper-
algesia. Although evaluated by antineuropathic drugs, further
study is needed to understand the induction of the major
pathogenesis of PDN.
2.4. Models Developed during 2000s
2.4.1. Nonobese Diabetic (NOD) Mice Models. Diabetic auto-
nomic neuropathy has been examined in the nonobese
diabetic (NOD), and streptozotocin (STZ)-induced diabetic
mice, two models of Type  diabetes, and the db/db mouse,

 Journal of Diabetes Research
amodelofTypediabetes[]. It was found that aer only –
 weeks of diabetes, NOD mice developed markedly swollen
axons and dendrites (neurotic dystrophy) in the prevertebral
superior mesenteric and celiac ganglia (SMG-CG), similar
to the pathology described in diabetic STZ- and BBW-rat
and human. STZ-induced diabetic mice develop identical
changes, although at a much slower pace and to a lesser degree
than NOD mice. Chronically diabetic Type  db/db mice
fail to develop neurotic dystrophy, suggesting that hypergly-
caemia alone may not be the crucial and sucient element.
erefore, NOD mouse appears to be a valuable model of dia-
betic sympathetic autonomic neuropathy which is consistent
with the pathogenesis of other rodent models and human. It
has been further supported by a very recently published com-
parative study on peripheral neuropathy between NOD and
ICR diabetic mice []whereNODmicehavebeensuggested
as a better model than ICR mice particularly in terms of nerve
regeneration.
2.4.2. Genetic Rodent Models. e development of peripheral
diabetic neuropathy has been assessed by longitudinal mem-
ory performance in spontaneously induced Type  diabetic
InsCY Akita mice by Choeiri et al. []. is model was
characterized by reduced number of beta cells with hypoinsu-
linemia, progressive hyperglycemia, and reduced sensory
nerve conduction velocity; however no signicant decit has
been detected as Morris water maze trial compared to the
control group, and many other diabetic neuropathy-related
major parameters have not been measured. Later, aer mea-
suring a number of diabetic neuropathy related parameters,
Schmidt et al. []reportedthatInsAkitamouseisarobust
model of diabetic sympathetic autonomic neuropathy which
closely corresponds to the characteristics pathology of other
rodent models and humans. is model has been evaluated
by progressively developed markedly swollen axons and
dendrites which are the common signs of neurotic dystrophy.
According to the above-mentioned studies, although Ins
Akita mice can be a proper genetic model of diabetic neu-
ropathy, this model needs to be evaluated by antidiabetic and
antineuropathic drugs.
Drel et al. [] reported that leptin-decient (ob/ob) mice
clearly manifest thermal hypoalgesia, the condition observed
in human subjects, which is a transient phenomenon in PDN
in humans [] and, non-fasting blood glucose was not more
than  mmol/L which was found very higher, ∼ mmol/L,
in Zucker Diabetic Fatty (ZDF) rats []. e ob/ob mice
developedaclearlymanifestedslowmotorandsensorynerve
conduction and accumulation of peripheral nerve sorbitol
pathway intermediate when fed a regular mouse diet to
maintain moderated hyperglycaemia []. Usually subject
with Type  or Type  diabetes display epidermal nerve ber
loss, and it was found that -week-old ob/ob mice developed
a dramatic reduction (%) in intraepidermal nerve ber
compared with age-matched nondiabetic controls []. is
animal model was also successfully evaluated by a potent
inhibitor of PDN such as aldose reductase inhibitor which
normalized motor and sensory nerve conduction velocity.
e results of this study suggest that leptin-decient ob/ob
mice can be better for PDN.
On the other hand, Kamenov et al. []comparedthe
complications of diabetic neuropathy between Otsuka Long-
Evans Tokushima Fatty (OLETF) rats and Long-Evans Tok-
ushima Otsuka (LETO) rats, where OLETF is a spontaneous
animal model of TD. In this regard, each type of animal
has been divided into  subgroups and fed with or without
sucrose-containing diets for  months and found that the
blood glucose and HbAc levels were signicantly higher in
OLETF rats, when compared with those in control LETO rats.
Motor nerve conduction velocity and thermal nociception
were signicantly decreased in OLETF rats in their  months
of age, while the values of the tail pressure test did not dier
compared with those from LETO rats. It was concluded that
signs of diabetic neuropathy appear in LETO rats aer a
longer period of time compared to OLETF rats. erefore
OLETF rat can be a better animal model for Type  diabetic
neuropathy than the LETO rats.
Recently, Seraf
´
ın et al. [] developed a model of diabetic
neuropathy in -week-old rat insulin I promoter/human
interferon-beta (RIP/IFN𝛽)transgenicICRmicewithalow
dose of STZ injection ( mg/kg BW) for  consecutive days.
Additionally, in order to induce nerve damage, aer  weeks
of sustained hyperglycemia, the le sciatic nerve was exposed
bybluntdissectionandcrushedatthefemurmajortrochanter
level for three times in succession for  seconds in anaes-
thetized animals when intact contralateral nerve was used
as a control. is transgenic model was evaluated by signif-
icant hyperglycemia, slower tibial sensory nerve conduction
velocity (SNCV) and increased motor latencies and duration
of compound muscle potential, reduced nerve ber density,
and so on. e slower recovery of nerve conduction velocities
were observed in the diabetic transgenic mice group com-
pared to the control. Although this model has been displayed
most of the major pathogenesis of peripheral diabetic neu-
ropathy, a sophisticated surgical approach has been used with
multipleSTZinjectionstodevelopthismodel,andithasnot
been evaluated by any antidiabetic or antineuropathic drugs.
2.4.3. Experimentally-Induced Models. FilhoandFazan[]
developed a streptozotocin (STZ)-induced model of phrenic
nerve neuropathy in rats. Diabetes was induced by a single
injection of streptozotocin to penile vein, and higher blood
glucose level conrmed the diabetic state. Le and right
fascicular areas and diameter of the phrenic nerves were
signicantly decreased in the proximal segments and right
segments, respectively. e phrenic nerves of diabetic rats
showed smaller myelinated axon diameters compared to
controls. e 𝑔 ratio for diabetic rats was signicantly lower
than the controls when these changes have been restored
by the daily injection (s.c.) of insulin ( U/kg body weight).
Although this model has been evaluated by insulin, no anti-
neuropathic drug has been used for the evaluation of this
model.
Aer a year, Obrosova and colleagues []triedto
develop a neuropathy model in female CBL/J mice by
feeding high-fat diet for a -week period. is model was
characterized by the decit of motor and sensory nerve con-
ductions, tactile allodynia, and thermal hypoalgesia; however
intradermal nerve ber loss or axonal atrophy was absent in

Journal of Diabetes Research 
this model. Although plasma FFA and insulin concentrations
were increased and glucose tolerance was impaired, the frank
hyperglycemia was absent in this model. According to the
data,althoughthismodelcanbeusedforprediabetesand
obesity related neuropathy, it cannot be used for chronic dia-
beticneuropathy.ismodelhasalsonotbeenevaluatedby
any antineuropathic drug, and the duration of model devel-
opment time is one of the major concerns.
In , Hong and Kang []publishedaveryspecial
nding on auditory neuropathy in streptozotocin-induced
diabetic ICR mice in order to understand the possible
auditory damage. e diabetes was induced by the dierent
dosages of STZ (, , and  mg/kg BW) dissolved in cit-
rate buer (pH .) in -week-old male animals. e auditory
diabetic neuropathy in this particular model has been evalu-
ated by signicantly increased absolute latencies of IV, and the
interpeak latencies of I–III and I–IV of auditory brainstem
response (ABR), and dose dependent induction of Pa latency
of auditory middle latency response (AMLR) in STZ treated
mice compared to control mice. In terms of ABR, best results
were observed for the dose of  mg/kg BW of STZ com-
pared to other two STZ dosages. From the data of this study,
authors suggested that the STZ-induced mouse can be used
for the evaluation of auditory pathway impairment via ABR
and AMLR tests, however this model has not been evaluated
by any antidiabetic or antineuropathic drugs.
Atthesameyear,Vareniuketal.[] compared the patho-
genesisofperipheraldiabeticneuropathyinSTZ-induced
wild-type and inducible nitric oxide synthase (iNOS) gene
decient mice with CBL/J background. e model was
developed by injecting single doses ( mg/kg BW) of STZ
injection (i.p.) to nonfasted wild-type and iNOS (also known
as Nos) decient (iNos (−/−)) mice and maintained for a
-week experimental period. Although STZ-injected wild-
type mice displayed peroxynitrite injury in peripheral nerve
and dorsal root ganglion neurons and developed motor and
sensory nerve conduction velocity decits, thermal and
mechanical hypoalgesia, tactile allodynia, and approximately
% loss of intraepidermal nerve bers, the STZ-injected
iNOS (−/−) mice did not display most of the above-men-
tioned pathogenesis except nitrosative stress in dorsal root
ganglia with normal nerve conduction velocities and less
severe small ber sensory neuropathy. Although the STZ
injected model was not evaluated by any antidiabetic or
antineuropathic drugs, but from this study it is clear that
iNOS gene plays a major role in the induction or peripheral
diabetic neuropathy which can be future research and drug
development target.
Recently, Muthuraman and colleagues []developeda
rat model of vasculatic neuropathy by ischemic perfusion in
the rat femoral artery. is model was validated aer , , and
 h of ischemia followed by prolonged reperfusion. e model
has been characterized by thermal and mechanical hyperal-
gesia in paw and tail which are associated with peripheral
andcentralneuropathicpain,respectively.eserumIL-,
nerve ber density, and nerve conduction velocity were lower,
and serum nitrate, malondialdehyde (MDA) and TNF-alpha
levels were higher in this model. Although neuropathy
induction period of this model is very short and has similar
pathogenesis with human diabetic neuropathy, the pathogen-
esis of neuropathy have not been developed here via hypergly-
caemia, what is usually happened in diabetic neuropathy, but
via ischemic perfusion in the animal femoral artery. Hence,
this model cannot be a better model to study human periph-
eral diabetic neuropathy. Additionally, this model has not
been evaluated by using any antineuropathic drugs.
2.5. Models Developed during 2010s
2.5.1. Genetically Modied SDT Rat Model. Recently, Yam-
aguchi et al. [] developed diabetic peripheral neuropathy in
Spontaneously Diabetic Torii (SDT) fatty rats by introducing
fa allele of Zucker Diabetic Fatty (ZDF) rats since SDT rats
develop delayed hyperglycemia compared to diabetic com-
plications. Apart from common diabetic abnormalities such
as sustained hyperglycemia and dyslipidemia, this diabetic
peripheral neuropathic model was further characterized by
signicantly delayed and lower motor nerve conduction
velocity from  weeks and signicantly lower number of
sural nerve bers at the end of the -week experimental
period. Additionally, thickened epineurial arterioles were
frequently found in this model. is model was further evalu-
ated by an antidiabetic drug such as pioglitazone which could
signicantly improve the motor nerve conduction velocity
and blood HbAc level when fed food admixture at a dose of
 mg/kg/day for a -week period. So this model can be a bet-
ter diabetic peripheral neuropathic model not only to under-
stand the pathogenesis of diabetic peripheral neuropathy but
also to screen and develop antidiabetic peripheral neuro-
pathic drug, particularly for Type  diabetes.
2.5.2. Genetically Modied C57BLKS Mice Model. Ver y
recently, Hinder et al. [] developed a dyslipidemia-induced
mouse model of diabetic neuropathy by some genetic manip-
ulation. is model was developed by knockout of ApoE and
ApoB genes in db/db or ob/ob mice CBLKS background
which mimicked the neuropathic plasma lipid prole in
diabetic humans. It was also characterized by increased body
weight, hyperlipidemia, hyperglycemia, and the evidence
of neuropathy; however this model was not delivered by
lipid prole usually seen in translational diabetic neuropathy.
Although this model has been characterized by signicantly
lowertailickresponsetoheatstimulus,sciaticmotornerve
conduction velocity, and intraepididymal nerve ber velocity,
mismatched results were observed for the body weight, blood
glucose, plasma lipids, and total blood glycated haemoglobin.
From the results of this study, authors suggested that the
overall eects of ApoE knockout, either directly upon nerve
structure and function or indirectly on lipid metabolism, are
insucient to signicantly alter the course of translational
diabetic neuropathy research, and further therapeutic inter-
vention is necessary in this regard. Apart from the above
limitations, this model was also not evaluated by any antidi-
abetic or antineuropathic drug.
2.5.3. Streptozotocin-Induced Diabetic Sensory Neuropathy
Mice Model. Most recently, Murakami et al. []developed

 Journal of Diabetes Research
a sensory neuropathy model in STZ-induced -week-old ddY
mice. Diabetes was developed by a single injection (i.p.) of
STZandconrmedbybloodglucoselevel>. mmol/L one
week aer the STZ injection. is model has been evaluated
by signicantly lower sensory nerve conduction velocity
(SNCV), higher nociceptive threshold, hypoalgesia, and
reducedaxonareaofunmyelinatednervebersorunmyeli-
nated ber atrophy. Although no dierence was found for the
myelinated nerve ber areas between the diabetic and healthy
mice, this model has been successfully evaluated by insulin
treatment. Since the unmyelinated nerve bers were more
aected than myelinated nerve bers and it has been success-
fully evaluated with insulin treatment, so it can be a better
modeltostudythehumansensorypolyneuropathy.
3. Conclusion
Asperthisreview,althoughanumberofapproacheshave
been used to develop the diabetic neuropathic models in dif-
ferent strains of animals in last ve decades, none of them are
without limitations. Several models such as conventional and
genetically modied CBL/Ks (db/db) mice, streptozotocin-
induced CBL/J and ddY mice, spontaneously diabetic
WBN/Kob rats, L-fucose induced neuropathic rats, nonobese
diabetic (NOD) rats, spontaneously induced InS Akita
mice, leptin-decient (ob/ob) mice, high-fat diet-fed female
CBL/J mice, and genetically modied SDT fatty rats have
been shown to develop major pathogenesis of diabetic neu-
ropathy or peripheral diabetic neuropathy; however most of
them were not validated either by antidiabetic or antineuro-
pathic drugs. Some models such as streptozotocin-induced
rats, Chinese hamster, rhesus monkey, partial sciatic nerve
ligated rats, and Otsuka Long-Evans Tokushima Fatty
(OLETF) rats developed very few major or some minor
pathogenesis of diabetic neuropathy and peripheral diabetic
neuropathy and the model development time for some of
these models were very long. e best model of diabetic neu-
ropathy or peripheral diabetic neuropathy should have some
major criteria such as: () the model should have all major
pathogenesis of diabetic neuropathy or PDN with other
minor pathogenesis which is normally found in human dia-
betic neuropathic patients, () the model should be sensitive
to antidiabetic or anti-neuropathic drugs, and () the model
needstobesuitabletostudythepathogenesisofdisease
as well as for routine pharmacological screening of antidia-
betic anti-neuropathic drugs. Although most of the genetic
or genetically modied models of diabetic neuropathy or
PDN discussed in this review are suitable for studying the
pathogenesis of the diseases, the CBL/Ks (db/db) mice,
streptozotocin-induced CBL/J and ddY mice, sponta-
neously diabetic WBN/Kob rats, nonobese diabetic mice,
spontaneously induced Ins Akita mice, and leptin-decient
(ob/ob)micehavebeenfoundasbettermodelsforhuman
diabetic neuropathy when high-fat diet-fed female CBL/J
mice have been suggested to use for prediabetic or obesity
related diabetic neuropathy. Although L-fucose induced neu-
ropathic rats, OLETF rats, and genetically modied SDT rats
have shown some promising pathogenesis of diabetic and
PDN, further studies are needed to understand the suitability
and usefulness of these models for diabetic or peripheral dia-
betic neuropathic researches.
List of Abbreviations (in Alphabetical Order)
DN: Diabetic neuropathy
GDM: Gestational diabetes mellitus
ICD: International classication of diseases
IDDM: Insulin dependent diabetes mellitus
IFN: Interferon
IGT: Impaired glucose tolerance
IND: International nomenclature of diseases
iNOS: Inducible nitric oxide synthase
LETO: Long Evans Tokushima obese
MNCV: Motor nerve conduction velocity
MRDM: Malnutrition related diabetes mellitus
NIDDM: Noninsulin dependent diabetes mellitus
NOD: Nonobese diabetic
OLETF: Otsuka long Evans Tokushima fatty
PDN: Peripheral diabetic neuropathy
SDT: Spontaneously diabetic torii
SNCV: Sensory nerve conduction velocity
STZ: Streptozotocin
TD: Type  diabetes
WHO: World health organization
ZDF: Zucker diabetic fatty.
Acknowledgments
is work was supported by a Competitive Research Grant
from Research Oce of the University of KwaZulu-Natal,
Durban and an Incentive Grant for Rated Researchers and a
Grant Support for Women and Young researchers from the
National Research Foundation (NRF), Pretoria, South Africa.
References
[] A. M. Ahmed, “History of diabetes mellitus,” Saudi Medical
Journal,vol.,no.,pp.–,.
[] WHO Expert Committee on Diabetes Mellitus. Second Report,
vol.  of WHO Technical Report Series, Geneva, Switzerland,
.
[] World Health Organization (WHO), DiAbetes Mellitus: Report
of a WHO Study Group,vol.ofWHO Technical Report Series,
Geneva, Switzerland, .
[] Y. Harati, “Diabetic neuropathies: unanswered questions,” Neu-
rologic Clinics,vol.,no.,pp.–,.
[] A.J.M.Boulton,“ediabeticfoot:fromarttoscience.e
th Camillo Golgi lecture,” Diabetologia,vol.,no.,pp.–
, .
[] V. R. Drel, N. Mashtalir, O. Ilnytska et al., “e leptin-decient
(ob/ob) mouse: a new animal model of peripheral neuropathy of
type  diabetes and obesity,” Diabetes, vol. , no. , pp. –
, .
[] N. A. Calcutt, “Potential mechanisms of neuropathic pain in
diabetes,” International Review of Neurobiology,vol.,pp.–
, .
[] K.A.Sullivan,S.I.Lentz,J.L.RobertsJr.,andE.L.Feldman,
“Criteria for creating and assessing mouse models of diabetic
neuropathy,” Current Drug Targets,vol.,no.,pp.–,.

Journal of Diabetes Research 
[] A. S. Shaikh and R. S. Somani, “Animal models and biomarkers
of neuropathy in diabetic rodents,” Indian Journal of Pharma-
cology,vol.,no.,pp.–,.
[] A. H
¨
oke, “Animal models of peripheral neuropathies,” Neu-
rotherapeutics,vol.,no.,pp.–,.
[] S. G. Eliasson, “Nerve conduction changes in experimental dia-
betes,” e Journal of Clinical Investigation,vol.,pp.–
, .
[] S. G. Eliasson, “Regenerative processes in experimental diabetic
neuropathy,” Transactions of the American Neurological Associ-
ation,vol.,pp.–,.
[] G. M. Preston, “Peripheral neuropathy in the alloxan-diabetic
rat,” Journal of Physiology,vol.,no.,.
[] R. E. Lovelace, “Experimental neuropathy in rats made diabetic
with alloxan,” Electroencephalography and Clinical Neurophysi-
ology,vol.,no.,p.,.
[] J. Jakobsen and K. Lundbeck, “Neuropathy in experimental dia-
betes: an animal model,” British Medical Journal,vol.,no.,
pp. –, .
[] A. A. F. Sima and D. M. Robertson, “Peripheral neuropathy in
mutant diabetic mouse [CBL/Ks(db/db)],” Acta Neuropatho-
logica,vol.,no.,pp.–,.
[] A. A. F. Sima and D. M. Robertson, “Peripheral neuropathy in
the diabetic mutant mouse. An ultrastructural study,” Labora-
tory Investigation,vol.,no.,pp.–,.
[] D. M. Robertson and A. A. F. Sima, “Diabetic neuropathy in
the mutant mouse [CBL/ks(db/db)]. A morphometric study,”
Diabetes,vol.,no.,pp.–,.
[] W. R. Kennedy, D. C. Quick, T. Miyoshi, and G. C. Gerritsen,
“Peripheral neurology of the diabetic Chinese hamster,” Dia-
betologia,vol.,no.,pp.–,.
[] D. R. Cornblath, M. A. Hillman, J. S. Strier, C. N. Herman,
andB.C.Hansen,“Peripheralneuropathyindiabeticmonkeys,”
Diabetes,vol.,no.,pp.–,.
[] A. A. F. Sima, W.-X. Zhang, and D. A. Greene, “Diabetic and
hypoglycemic neuropathy—a comparison in the BB rat,” Dia-
betes Research and Clinical Practice,vol.,no.,pp.–,
.
[] O. A. R. Filho and V. P. S. Fazan, “Streptozotocin induced dia-
betes as a model of phrenic nerve neuropathy in rats,” Journal of
Neuroscience Methods,vol.,no.,pp.–,.
[] L. M. Hinder, A. M. Vincent, J. M. Hayes, L. L. McLean, and E.
L. Feldman, “Apolipoprotein E knockout as the basis of mouse
models of dyslipidemia-induced neuropathy,” Experimental
Neurology, vol. , pp. –, .
[] I. Vareniuk, I. A. Pavlov, and I. G. Obrosova, “Inducible nitric
oxide synthase gene deciency counteracts multiple manifes-
tations of peripheral neuropathy in a streptozotocin-induced
mouse model of diabetes,” Diabetologia, vol. , no. , pp. –
, .
[] T. Murakami, T. Iwanaga, Y. Ogawa et al., “Development of
sensory neuropathy in streptozotocin-induced diabetic mice,”
Brain and Behavior,vol.,no.,pp.–,.
[] S. Yagihashi, R.-I. Wada, M. Kamijo, and K. Nagai, “Peripheral
neuropathy in the WBN/Kob rat with chronic pancreatitis and
spontaneous diabetes,” Laboratory Investigation,vol.,no.,
pp. –, .
[] A. A. F. Sima, J. A. Dunlap, E. P. Davidson et al., “Supplemental
myo-inositol prevents L-fucose-induced diabetic neuropathy,”
Diabetes, vol. , no. , pp. –, .
[] A. Fox, C. Eastwood, C. Gentry, D. Manning, and L. Urban,
“Critical evaluation of the streptozotocin model of painful
diabetic neuropathy in the rat,” Pain,vol.,no.,pp.–,
.
[] R. E. Schmidt, D. A. Dorsey, L. N. Beaudet et al., “Non-obese
diabetic mice rapidly develop dramatic sympathetic neuritic
dystrophy: a new experimental model of diabetic autonomic
neuropathy,” American Journal of Pathology,vol.,no.,pp.
–, .
[] J. Homs, L. Ariza, G. Pag
`
es et al., “Comparative study of
peripheral neuropathy and nerve regeneration in NOD and ICR
diabetic mice,” Journal of the Peripheral Nervous System,vol.,
no.,pp.–,.
[] C. Choeiri, K. Hewitt, J. Durkin, C. J. Simard, J.-M. Renaud,
and C. Messier, “Longitudinal evaluation of memory perfor-
mance and peripheral neuropathy in the InsCY Akita mice,”
Behavioural Brain Research,vol.,no.,pp.–,.
[] R. E. Schmidt, K. G. Green, L. L. Snipes, and D. Feng, “Neuritic
dystrophy and neuronopathy in Akita (InsAkita) diabetic
mouse sympathetic ganglia,” Experimental Neurology,vol.,
no. , pp. –, .
[] Z. Kamenov, H. Higashino, M. Todorova, N. Kajimoto, and A.
Suzuki, “Physiological characteristics of diabetic neuropathy in
sucrose-fed Otsuka Long-Evans Tokushima fatty rats,” Methods
and Findings in Experimental and Clinical Pharmacology,vol.
,no.,pp.–,.
[] A. Seraf
´
ın, J. Mol
´
ın, M. M
´
arquez et al., “Diabetic neuropa-
thy: electrophysiologicaland morphological study of peripheral
nerve degenerationand regeneration in transgenic mice that
express IFN𝛽 in 𝛽 cells,” Muscle and Nerve,vol.,no.,pp.
–, .
[] I. G. Obrosova, O. Ilnytska, V. V. Lyzogubov et al., “High-fat
diet-induced neuropathy of pre-diabetes and obesity: eects of
“healthy” diet and aldose reductase inhibition,” Diabetes,vol.,
no. , pp. –, .
[] A. Muthuraman, M. Ramesh, and S. Sood, “Development
of animal model for vasculatic neuropathy: induction by is-
chemic-reperfusion in the rat femoral artery,” Journal of Neu-
roscience Methods,vol.,no.,pp.–,.
[] T. Yamaguchi, T. Sasase, Y. Mera et al., “Diabetic peripheral
neuropathy in spontaneously diabetic tori-Lepr (fa) (SDT Faty)
rats,” Journal of Veterinary Medical Science,vol.,no.,pp.
–, .
[] K. Ozaki, K. Miura, M. Tsuchitani, and I. Narama, “Peripheral
neuropathy in the spontaneously diabetic WBN/Kob rat,” Acta
Neuropathologica,vol.,no.,pp.–,.
[] P. J. Dyck, P. J. B. Dyck, J. A. Velosa, T. S. Larson, and P. C.
O’Brien, “Patterns of quantitative sensation testing of hypoes-
thesia and hyperalgesia are predictive of diabetic polyneuropa-
thy: a study of three cohorts,” Diabetes Care,vol.,no.,pp.
–, .
[] B. N. Hong and T. H. Kang, “Auditory neuropathy in strepto-
zotocin-induced diabetic mouse,” Neuroscience Letters,vol.,
no.,pp.–,.